Langmuir 2001, 17, 1069-1073
1069
Flocculation of Latex Particles of Varying Surface Charge Densities by Chitosans Matthew Ashmore, John Hearn,* and Franciszek Karpowicz Department of Chemistry and Physics, Nottingham Trent University, Clifton Lane, Nottingham NG11 8NS, United Kingdom Received October 23, 2000. In Final Form: December 8, 2000 A series of well-characterized chitosan samples spanning a wide range of molecular weights, having been homogeneously acetylated to produce varying degrees of linear charge density, were used to flocculate “model colloid” polymer latices. Latices were selected to cover a wide range of charge densities and were sized between 85 nm and 2.1 µm. The optimum flocculation concentrations were identified using both residual turbidity and initial rate methods together with electrophoretic mobility measurements. A charge neutralization mechanism with an enhancement of rate in comparison to the rate of rapid coagulation indicative of a “charge patch” effect was confirmed, except in the case of the 2.1 µm particles. The optimum flocculation concentration of the latter fell below that of rapid coagulation, an effect ascribed to the influence of adsorbed chitosan upon the drainage between particles on close approach. Both molecular weight and degree of acetylation of the chitosans had a small effect upon the optimum flocculation concentration, with efficiency increasing with increases in both molecular weight and linear charge density. Increasing the ionic strength of the dispersion medium broadened the flocculation concentration range and diminished the enhancement of rate produced by the charge patch mechanism. With doses of chitosans in excess of the optimum for flocculation, charge reversal occurred and the cationic latices were rather more stable to electrolyte addition than their anionic precursors but were electrostatically rather than sterically stabilized.
Introduction
Experimental Section
Chitosan, a biodegradable, nontoxic, and renewable resource commodity, is a naturally occurring copolymer of poly[β(1f 4)-2-acetamido-2-deoxy-D-glucopyranose] and poly[β(1f4)-2-amino-2-deoxy-D-glucopyranose], which becomes soluble and positively charged in acidic media and may therefore be used as an environmentally friendly flocculant.1 The linear charge density depends on the level of acetylation. The mechanism by which chitosans flocculate particulate dispersions has been a matter of debate and disagreement in the published literature,2-7 often because the systems used have been insufficiently welldefined, particularly in large scale applications.8-10 A recent publication11 has addressed this problem by reporting results for a series of well-characterized chitosans with a single particle size model colloid latex. This paper reports the effects of using well-characterized polystyrene latices of differing surface charge densities in studies of flocculation by chitosans of differing linear charge densities and molecular weights.
Materials. All water used was doubly distilled from an allPyrex apparatus. Acetic acid was GPR grade (>99.5%) supplied by BDH, Poole, Dorset, U.K. Styrene (>99%) was supplied by Sigma Aldrich, Poole, Dorset, U.K. and was distilled under reduced pressure of nitrogen gas to remove the stabilizer. Sodium dodecyl sulfate (SDS) (>98%) was supplied by Fluka, Gillingham, Dorset, U.K. Potassium persulfate (KPS) SLR grade (>97%) recrystallized from distilled water, cetyltrimethylammonium bromide (CTAB) AR grade (>99%), sodium chloride SLR grade (99.6% after drying), hydrochloric acid SLR, and carbonate-free sodium hydroxide for volumetric analysis were all supplied by Fischer Scientific, Loughborough, Leicestershire, U.K. In the preparation and use of solutions of cationic materials, all vessels were conditioned to allow for losses by adsorption onto the walls;12 that is, the vessels were first equilibrated over a period of 12 h with a solution of the desired concentration which was then discarded and replaced with a new solution at the desired concentration. Chitosans. Chitosan samples are coded according to their fractional acetylation FA and their molecular weight; for example, FA[0.02]O is 2% acetylated oligomeric (O) (relative molecular mass (RMM) ca. 2 × 103) chitosan. Medium molecular weight (M) (RMM ca. 5 × 105) and high molecular weight (H) (RMM ca. 1 × 106) samples were also used and were homogeneously reacetylated to increasing degrees which were then quantified by dye adsorption.11 Latex Particles. Surfactant-free polystyrene latex particles, stabilized solely by surface sulfate groups and spanning the full range of surface charge densities available, were obtained from the Interfacial Dynamics Corp., Portland, OR as follows: 350 nm diameter, 0.33 µC cm-2 (product no. 1-300, batch no. 4011); 400 nm diameter, 3.38 µC cm-2 (product no. 1-400, batch no. 2-63s44.186,1); 2.1 µm diameter, 8.6 µC cm-2 (product no. 1-2000, batch no. 780). Latex particles (85 nm, coefficient of variance 6%) were prepared by emulsion polymerization of styrene at 40 °C using SDS emulsifier at 10× its critical micelle concentration, and the initiator was KPS. The latex particles were cleaned by exhaustive dialysis, acid washed with 0.01 M HCl, and then serum exchanged
(1) Roberts, G. A. F. In Adv. Chitin Sci. Vol. II.; Proc. 7th International Conf. Chitin & Chitosan; Jacques Andre: Lyons, 1997; p 22. (2) Zhou, Y.; Huqun, L.; Shoucheng, D.; Weixia, Z.; Gonghui, Z. Wuli Huaxue Xuebao 1993, 9 (1), 77. (3) Agervist, I. Colloids Surf. 1992, 69 (2-3), 173. (4) Huang, C.; Chen, Y. J. Chem. Technol. Biotechnol. 1996, 66, 227. (5) Ordolff, D. Kiel. Milchwirstsch. Forshungsber. 1995, 47 (4), 339. (6) Pinotti, A.; Bevilacqua, A.; Zaritzky, N. J. Food Eng. 1997, 32 (1), 69. (7) Domard, A.; Rinaudo, M.; Terrassin, C. J. Appl. Polym. Sci. 1989, 38, 1799. (8) Wu, A. C. M.; Brough, W. A. In Proc. 1st International Conf. Chitin & Chitosan, 1977; Muzzarelli, R. A. A., Pariser, E. R., Eds.; MIT Sea Grant Program Report MITSG 78-7, 1978; p 88. (9) Ganjidoust, H.; Tatsumi, K.; Yamagishi, T.; Gholian, R. N. Water Sci. Technol. 1997, 35 (2 & 3), 291. (10) Castellanos-Perez, N.; Maldonado-Vega, A.; Fernadez-Villagomez, G.; Caffarel-Mendez, S. In Chitin & Chitosan; Skjak-Braek, G., Anthonsen, T., Sandford, P. A., Eds.; Elsevier Applied Science: London, 1989; p 567. (11) Ashmore, M. H.; Hearn, J. Langmuir 2000, 16 (11), 4906.
(12) Gregory, J.; Sheiham, I. Br. Polym. J. 1974, 6, 47.
10.1021/la001490b CCC: $20.00 © 2001 American Chemical Society Published on Web 01/20/2001
1070
Langmuir, Vol. 17, No. 4, 2001
Ashmore et al.
with doubly distilled water until the filtrate had the conductivity of the distilled water. The concentration of the surface functional groups was determined by conductometric titration with carbonate-free sodium hydroxide, using established procedures.13 Flocculation Studies. Flocculation Rates. The initial rates of increase in turbidity of the 85, 350, and 400 nm polystyrene latex particles upon aggregation were estimated by following the increase in absorbance at 550 nm using a UNICAM 8720 UV/vis spectrophotometer in kinetics mode. The cuvette containing 2 mL of diluted stock latex dispersion (0.5 µm27-30 in diameter; thus, the effects of adsorbed chitosan on viscous drainage between particles may act to decrease the overall flocculation rate in the system. This effect would mask any enhancement of rate arising from charge patch interaction. When the 2.1 µm latex particles were restabilized as cationic particles by addition of chitosan FA[0.17]H at twice the optimum flocculation concentration, only a very small number of doublets (2.81% of the singlet population)23 were formed and the stability to electrolyte addition increased from 250 to 340 mM. This level of electrolyte tolerance suggests that the cationic particles were still charge stabilized rather than sterically stabilized when a critical coagulation concentration in excess of 5 M might be expected for a univalent electrolyte.31 This is a further (26) Overbeek, J. Th. G. In Colloid Chemistry; Kruyt, H. R., Ed.; Elsevier: Amsterdam, 1954; Vol. 1, p 278. (27) Higuchi, W. I.; Okada, R.; Stelter, G. A.; Lemberger, P. A. J. Pharm. Sci. 1963, 52, 49. (28) Smith, D. L.; Friedlander, S. K. J. Colloid Interface Sci. 1964, 19, 621. (29) Lichtenbelt, J. W. Th.; Ras, H. J. M. C.; Wiersema, P. H. J. Colloid Interface Sci. 1974, 46, 522. (30) Lichtenbelt, J. W. Th.; Pathmanoharan, C.; Wiersema, P. H. J. Colloid Interface Sci. 1974, 49, 281. (31) Buscall, R. In Polymer Colloids; Buscall, R., Corner, T., Stageman, J. F., Eds.; Elsevier Applied Science: London, 1985; p 197.
Flocculation of Latex Particles
indication that chitosan chains adopt a relatively flat configuration on the particle surface. Charge Separations. The charge patch mechanism relies upon an excess of charge on the flocculant chain compared with that on the particle surface in order to produce the oppositely charged patches which illicit an increase in aggregation rate over that of rapid coagulation. Important quantities to compare are the average distances between charges on the polyelectrolyte and on the particles. X-ray crystallography data suggest that the amide groups in chitin are ca. 1.5 nm apart,32 so the spacings between charges in nearly fully deacetylated chitosan would be expected to be similar. Table 2 shows the average spacing for chitosans homogeneously reacetylated to yield a random distribution of acetyl groups together with the measured rate enhancement for the 400 nm latex particles with an average spacing of 2.5 nm between anion groups. As the degree of acetylation of the polyelectrolyte increases, the average spacing between charges increases and the effectiveness of the patch produced is decreased; for example, chitosan FA[0.57]M at 2.7 nm has a separation greater than the intercharge distance on the latex particles, so the flocculation rate would not be expected to be much enhanced above that of simple coagulation, with any enhancement arising from an increased hydrophobic effect as found, for example, with CTAB. In comparing charge spacings on the comparably sized 350 and 450 nm latices, the former at the highest estimate of its surface charge has a spacing of 3.4 nm between charges as compared to 2.5 nm for the latter. The somewhat greater rate enhancement observed for the 350 nm latex particles at 45% as compared to 30% for the 400 nm latex particles reflects the expected greater cationic patch density for the latex particles with the lower surface charge density. Conclusions As identified by peak rate aggregation, minimum levels of residual turbidity, and electrophoresis results, the optimum flocculation concentration for all the latex particle sizes (85 nm, 350 nm, 400 nm, and 2.1 µm) occurs at or close to charge neutralization, in agreement with results obtained using CTAB. Greater concentrations of (32) Minke, R.; Blackwell, J. J. Mol. Biol. 1978, 120, 167.
Langmuir, Vol. 17, No. 4, 2001 1073
chitosans with lower linear charge densities are required to effect optimum flocculation, which occurs at or close to a 1:1 stoichiometric ratio of protonated amine groups to anionic charge on the latex particles. An enhancement of rate for optimum flocculation, in comparison to the rapid rate of coagulation, was found for all latices except the largest, 2.1 microns. The enhancement of rate is indicative of a charge patch mechanism of flocculation. In the case of the 2.1 µm latex particles, the optimum rate of flocculation was slower than the rate of rapid coagulation and this was ascribed to the influence of the adsorbed chitosan layer on drainage between the particles on close approach. Care was needed to avoid orthokinetic flocculation upon sampling in perikinetic experiments. For the other latices, an increasing ionic strength of the dispersion medium led to a reduction in the enhanced rate, caused by the charge patch effect, toward that of rapid coagulation. The increasing rigidity of the chitosan backbone chain with an increasing degree of acetylation, for the medium molecular weight chitosans, led to a slightly reduced flocculation efficiency at a little above stoichiometric. This was attributed to the increased difficulty for the chitosan to match the template of charges set by the latex particle surface. For a higher molecular weight chitosan, the effect was less noticeable and this was ascribed to the greater density of the charge patch attainable with the longer chain length molecule. Addition of chitosan concentrations above the optimum flocculation concentration led to charge reversal. The cationic latex particles then had a higher critical coagulation concentration to added electrolyte than their anionic precursors. They were, however, electrostatically stabilized rather than sterically stabilized by the adsorbed chitosans. Only negligible levels of doublet aggregates were formed upon charge reversal. Acknowledgment. The authors sincerely thank Professor George Roberts for providing chitosan samples and for advice on their characterization, Dr. Jim Goodwin, President of the Interfacial Dynamics Corporation, for providing titration data on latices, and Mr. Roland Crouch of Beckman Coulter for help in obtaining electrophoretic mobility and laser light scattering data. LA001490B
